Miriam A. M. Capretz

Yohannis Kifle Telila, Damitha Senevirathne, Dumindu Tissera et al.

Anomaly detection is crucial in the energy sector to identify irregular patterns indicating equipment failures, energy theft, or other issues. Machine learning techniques for anomaly detection have achieved great success, but are typically centralized, involving sharing local data with a central server which raises privacy and security concerns. Federated Learning (FL) has been gaining popularity as it enables distributed learning without sharing local data. However, FL depends on neural networks, which are vulnerable to adversarial attacks that manipulate data, leading models to make erroneous predictions. While adversarial attacks have been explored in the image domain, they remain largely unexplored in time series problems, especially in the energy domain. Moreover, the effect of adversarial attacks in the FL setting is also mostly unknown. This paper assesses the vulnerability of FL-based anomaly detection in energy data to adversarial attacks. Specifically, two state-of-the-art models, Long Short Term Memory (LSTM) and Transformers, are used to detect anomalies in an FL setting, and two white-box attack methods, Fast Gradient Sign Method (FGSM) and Projected Gradient Descent (PGD), are employed to perturb the data. The results show that FL is more sensitive to PGD attacks than to FGSM attacks, attributed to PGD's iterative nature, resulting in an accuracy drop of over 10% even with naive, weaker attacks. Moreover, FL is more affected by these attacks than centralized learning, highlighting the need for defense mechanisms in FL.

Norman Tasfi, Miriam Capretz

We introduce Dynamic Planning Networks (DPN), a novel architecture for deep reinforcement learning, that combines model-based and model-free aspects for online planning. Our architecture learns to dynamically construct plans using a learned state-transition model by selecting and traversing between simulated states and actions to maximize information before acting. In contrast to model-free methods, model-based planning lets the agent efficiently test action hypotheses without performing costly trial-and-error in the environment. DPN learns to efficiently form plans by expanding a single action-conditional state transition at a time instead of exhaustively evaluating each action, reducing the required number of state-transitions during planning by up to 96%. We observe various emergent planning patterns used to solve environments, including classical search methods such as breadth-first and depth-first search. DPN shows improved data efficiency, performance, and generalization to new and unseen domains in comparison to several baselines.

Michael Kernahan, Miriam Capretz, Luiz Fernando Capretz

The maintenance portion of the software lifecycle represents a major drain on most software companys resources. The transition from programmers to maintainers is high risk, since usually the maintainers have to learn the system from scratch before they can begin modifying it appropriately. This paper introduces a method for automatically extracting important traceability information from a C# software projects source code. Using this traceability information, maintainers (and programmers) are better able to evaluate the impacts their actions will have on the entire project.

Hany F. El Yamany, Miriam Capretz, Luiz Fernando Capretz

Software testing is a very expensive and time consuming process. It can account for up to 50% of the total cost of the software development. Distributed systems make software testing a daunting task. The research described in this paper investigates a novel multi-agent framework for testing 3-tier distributed systems. This paper describes the framework architecture as well as the communication mechanism among agents in the architecture. Web-based application is examined as a case study to validate the proposed framework. The framework is considered as a step forward to automate testing for distributed systems in order to enhance their reliability within an acceptable range of cost and time.

Michael Kernahan, Miriam Capretz, Luiz Fernando Capretz

Traceability information is a valuable asset that software development teams can leverage to minimise their risk during production and maintenance of software projects. When maintainers are added to a software project post-production, they have to learn the system from scratch and understand its dynamics before they can begin making appropriate modifications to the source code. The system outlined in this paper extracts traceability information directly from the source code of C# projects, and presents it in such a way that it can be easily used to understand the logic and validate changes to the system.